This study took pomegranate juice as the experimental subject and conducted single-factor and orthogonal experiments using chitosan, xanthan gum, guar gum, locust bean gum, and sodium alginate, with the aim of determining the optimal clarification process for pomegranate juice. Meanwhile, gas chromatography-mass spectrometry (GC-MS) technology was employed to analyze the volatile compounds in naturally clarified pomegranate juice and pomegranate juice treated with composite clarifying agents. In terms of evaluating clarification effects, the transmittance of pomegranate juice was used as an indicator to compare the clarification effects of the five clarifying agents. Through orthogonal optimization experiments, the optimal clarification process conditions for pomegranate juice were ultimately determined: A compound ratio of 0.1 g/L chitosan to 0.1 g/L sodium alginate at 1:1.5, a flocculation temperature of 30 ℃, and a calcium chloride addition amount of 0.05 g/L, achieving a transmittance of 82.357%. Additionally, the GC-MS results revealed that after treatment with the composite clarifying agent group, the relative content of ethyl acetate significantly increased, enhancing the fruity aroma. The relative content of 4-terpineol increased by 61.49%, adding woody, citrus, and floral notes. The relative content of 2-ethylhexanol increased by 34.79%, contributing sweet and fruity aromas. The relative content of nonanal rose by 54.75%. Nonanal possessed a fresh citrus peel aroma, which helped retain a mild aldehyde flavor and stabilized the overall flavor profile. Moreover, the composite clarifying agent group reduced the relative contents of ethanol, (+)-α-terpineol, n-hexanal, trans,trans-2,4-nonadienal, and other substances that might produce undesirable flavors, thereby decreasing the pungent odor and rancid oil smell in pomegranate juice and further improving its overall flavor. This study systematically explores the impact of processing techniques on the quality of pomegranate juice, and its research findings can provide a theoretical basis for improving the sensory and nutritional quality of pomegranate juice products.

To establish a sustainable vegetative propagation system for U. pinnatifida gametophytes and enhance mycosporine–likeamino acids (MAAs) accumulation, this study systematically investigated the effects of key culture conditions—such as light quality, light intensity, photoperiod, temperature, and nutrient availability—on gametophyte growth and MAAs accumulation using a combination of single-factor and uniform design experiments. The antioxidant activity of MAAs extract was evaluated in vitro via DPPH and ABTS⁺ radical scavenging assays, while their anti-inflammatory potential was assessed in vivo using a zebrafish inflammation model. Results indicated that culture conditions significantly affected both gametophyte growth and MAAs content. Optimal conditions—red light, 20 ℃, a light intensity of 32 μmol·m⁻²·s⁻¹, and a 12L:12D photoperiod—yielded the highest relative growth rate (6.41%±0.11%) and MAAs content (327.19±1.59 mg/g). UV-visible spectroscopy confirmed the characteristic absorption profiles of MAAs in the extracts. The MAAs extract exhibited IC₅₀ values of 12.04 and 10.34 mg/mL against DPPH and ABTS⁺ radicals, respectively. Furthermore, at 100 μg/mL, the extract reduced macrophage aggregation in zebrafish by 52.31%, demonstrating a clear dose-dependent anti-inflammatory effect. In summary, this work identifies optimal culture conditions for enhancing both gametophyte growth and MAAs production, and confirms that gametophyte-derived MAAs possess moderate antioxidant and notable anti-inflammatory activities, thereby providing a theoretical and experimental foundation for their efficient biosynthesis and potential application.

To establish an eco-friendly and efficient method for extracting natural lutein while enhancing its stability, this study employed menthol-based deep eutectic solvents (DESs) for the extraction of lutein from chrysanthemum and the preparation of lutein-menthol cocrystals. The feasibility of this approach was evaluated through the analysis of the antioxidant activity of the extracted lutein solutions, as well as the stability and composition of the cocrystals. Firstly, the extraction efficiency of six DESs (menthol-acetic acid, menthol-propionic acid, menthol-octanoic acid, menthol-decanoic acid, menthol-hexanoic acid, and menthol-lactic acid) for lutein was compared, identifying the optimal eutectic solvent. Subsequently, single-factor investigations were conducted on ethanol content, solid-to-liquid ratio, extraction temperature, and extraction time, and the extraction process parameters were optimized by using the response surface method. At the same time, the antioxidant activity of the extracted lutein solution was evaluated, and the differences in light and heat stability between the lutein-piperonal complex and the commercially available lutein powder were investigated. The cocrystal was analyzed using liquid chromatography-tandem mass spectrometry (LC-MS/MS). The results indicated that menthol-acetic acid (DES-1) was the optimal eutectic solvent; the optimized extraction conditions were 47% (V/V) ethanol content, a material-to-liquid ratio of 1:45 (g/mL), and an extraction temperature of 69 ℃, under which the extraction yield of lutein can reached 644.11±1.99 μg/g. The lutein solution extracted and prepared with DES-1 exhibits scavenging rates of 77.52% and 84.58% for hydroxyl radicals and DPPH radicals, respectively, which were significantly higher than those achieved by traditional ethanol extraction methods (P<0.01). Under thermal conditions of 30 ℃ at 75% relative humidity and light conditions of 4500 Lux at 25 ℃, the lutein retention rates for the lutein-menthol cocrystal were 84.34% and 33.73%, both of which were significantly higher than those of commercially available lutein. The formed cocrystal structure could physically hinder oxygen penetration, while the hydroxyl groups at both ends of lutein can form strong hydrogen bonds with the hydroxyl groups of menthol, thereby significantly enhancing stability. Identification through LC-MS/MS revealed a total of 33 lutein-related components in the cocrystal, including lutein and lutein palmitate, with the highest content of free lutein reaching 107.93±11.36 μg/g. The lutein-menthol cocrystal prepared in this study significantly enhances the stability of lutein and may provide more possibilities for the practical development and utilization of lutein.

Chestnut pollen, rich in high-value protein, was extensively discarded due to the lack of efficient extraction technologies, limiting its potential applications in functional foods and pharmaceuticals. This study presented the first systematic comparative analysis of three extraction methods—Ultrasound-assisted alkali dissolution and acid precipitation (UADAP), microwave-assisted alkali dissolution and acid precipitation (MADAP), and ammonium sulfate precipitation (ASP)—on the structural attributes and biological activities of chestnut pollen protein. With protein yield as the optimization criterion, the extraction process was systematically optimized through single-factor experiments integrated with response surface methodology. Results showed that all three methods yielded typical proteins with molecular weights of 20~100 kDa, exhibiting characteristic ultraviolet and infrared absorption peaks. Thermogravimetric analysis revealed thermal degradation at approximately 280 ℃, with MADAP-extracted protein demonstrating the highest thermal stability. Antioxidant activity assays indicated that MADAP protein exhibited superior activity in DPPH radical scavenging, hydroxyl radical scavenging, and ABTS⁺ radical scavenging rate, with an activity order of MADAP>UADAP>ASP. For hypoglycemic activity, the three chestnut pollen proteins all exhibited favorable inhibitory activities against α-glucosidase and α-amylase. Among them, MADAP showed the optimal inhibitory effect on α-glucosidase, UADAP possessed the strongest inhibitory activity against α-amylase, while ASP displayed the weakest inhibitory capacity toward both digestive enzymes. Response surface methodology optimized MADAP conditions (solid-to-liquid ratio 1:25 g/mL, extraction time 60 min, temperature 60 ℃), achieving a protein yield of 26.41%. In conclusion, the MADAP method effectively preserved protein structural stability while significantly enhancing its antioxidant and hypoglycemic activities, providing an optimal technical approach for the efficient development of chestnut pollen protein.

To enhance the in vitro antioxidant performance of raffinose, this study investigated the effects of temperature, time, pH, and the ratio of modifiers on the phosphate content in raffinose as single factors. The preparation process of raffinose was optimized by response surface methodology. The structures of raffinose before and after phosphorylation modification were characterized by Fourier transform infrared spectroscopy (FT-IR) and nuclear magnetic resonance spectroscopy (NMR), and their in vitro antioxidant activities were studied. The results showed that the optimal preparation process of phosphorylated raffinose (P-raffinose) was as follows: pH of the reaction system was 9, the ratio of sodium tripolyphosphate to sodium trimetaphosphate was 4:3, the reaction temperature was 77 ℃, and the reaction time was 5.3 h. Under these conditions, the phosphate content in P-raffinose was 32.14%. Characterization results indicated that raffinose was successfully phosphorylated. Activity tests demonstrated that phosphorylation modification significantly improved the hydroxyl radical scavenging capacity and total antioxidant capacity of raffinose, increasing by 27.06% and 31.40% respectively compared to raffinose. This study provides theoretical basis and technical support for the application of P-raffinose in the development of new functional foods and antioxidants.

This study employed artificial olfaction–vision equipment combined with multidata fusion and deep learning technologies to accurately predict the freshness of chilled beef, providing a theoretical basis and methodological support for the rapid detection of meat product freshness. In this study, a self-developed electronic nose and machine vision system was used to collect odor and image information from beef samples. The semi-micro Kjeldahl method and an acid–base meter were used to determine the total volatile basic nitrogen content and pH of the samples, which served as the basis for evaluating the freshness of the beef samples. Firstly, kernel principal component analysis (KPCA) was used to reduce the dimensionality of the olfactory–visual feature information of beef. Subsequently, prediction models for the electronic nose, machine vision, and data fusion were established using the backpropagation neural network (BPNN), support vector machine (SVM), convolutional neural network (CNN), and slime mold algorithm–convolutional neural network–bidirectional gated recurrent unit with attention (SMA–CNN–BiGRU-Attention) methods, and the model results were compared. The results indicated that the prediction performance of the data fusion model was significantly better than that of a single electronic nose or machine vision model. Among them, the SMA–CNN–BiGRU-Attention data-fusion model exhibited optimal stability, 100% accuracy in classifying the three freshness levels of beef, and 96.4% accuracy in classifying beef stored under chilled conditions for 12 days. This study demonstrates the feasibility of artificial olfaction–vision technology for detecting the freshness of meat products. The SMA–CNN–BiGRU-Attention model, based on data fusion, achieves high accuracy, offering a potential solution for detecting beef freshness.

The quality stability of fresh fruit milk tea is largely determined by the properties of its raw materials. As a commonly used ingredients, strawberries contribute significantly to the final flavor profile through their distinct aromas. Subtle differences in key aroma components can impart distinct aroma profiles to the final product, leading to batch-to-batch flavor variations in fresh fruit milk tea. Therefore, identifying the main aroma compounds in three strawberry cultivars commonly used in processing can help determine the causes of flavor differences. A combined approach of GC-MS (gas chromatography-mass spectrometry) and GC-IMS (gas chromatography-ion mobility spectrometry) was employed to separate and identify VOCs (volatile organic compounds) in the strawberries. OAV (odor activity value) analysis was utilized to pinpoint the key flavor compounds, and PCA (principal component analysis) combined with sensory evaluation was conducted to elucidate the aroma characteristic differences among the varieties. The results showed that a total of 110 VOCs were detected in 'Jiujiu' strawberry, 110 in 'Hani' strawberry, and 111 in 'Monterrey' strawberry. Through OAV analysis, 16 characteristic VOCs were identified, among which δ-decalactone, isoamyl acetate, linalool, (Z)-3-hexenal, (E,E)-2,4-nonadienal, 2,5-dimethyl-4-methoxy-3(2H)-furanone, ethyl isobutyrate, ethyl hexanoate, propionaldehyde, and 1-octen-3-ol were the key volatile components across all samples. PCA and sensory evaluation revealed significant differences in aroma characteristics among the three varieties, 'Jiujiu' strawberry exhibited strong sweet and floral aromas, attributed to the contributions of (Z)-3-hexenal and linalool, 'Hani' strawberry showed intense fruity and sour aromas, mainly derived from ethyl isobutyrate and ethyl hexanoate, while 'Monterrey' strawberry's fruity and fresh aromas were primarily influenced by ethyl isobutyrate and propionaldehyde. Additionally, differences in the quantitative levels of VOCs were also observed among the different strawberry cultivars. In summary, the combined GC-MS/GC-IMS technique clarified the key aroma substances and their differences across the three strawberries, and the established flavor fingerprint provided a theoretical basis for raw material selection in strawberry processing.

The aim of this study was to systematically investigate the volatile flavor profiles of goji berry puree and rehydrated products, and to identify key flavor markers of the puree. Using headspace-gas chromatography-ion mobility spectrometry (HS-GC-IMS), a total of 90 volatile compounds were detected, corresponding to 111 signal peaks, with some compounds occurring in monomeric, dimeric, or trimeric forms. Based on their chemical structures, these compounds were classified into nine categories: Alcohols, aldehydes, ketones, esters, heterocyclic compounds, terpenes, acids, sulfur-containing compounds, and hydrocarbons. Among these, aldehydes, alcohols, ketones, esters, and acids accounted for 87.11% to 95.51% of the total relative content. Through synergistic interactions, these compounds formed a complex and persistent aroma system that established the dominant floral, fruity, and sweet notes. Chemometric analysis was conducted on ten samples using SIMCA 14.1 software. Volatile flavor markers of goji berry puree were screened according to the following criteria: Variable importance in projection (VIP)>1.0, fold change (FC)>1.2 or <0.83, P<0.05, and coefficient>0. Ultimately, four characteristic volatile markers were identified in goji berry puree: Thiophene, acetone, 2-butanone, and furfural (monomer). Regression coefficient analysis indicated that goji berry puree was primarily characterized by grassy, fruity, and vegetable-like flavor attributes, while rehydrated puree exhibited more pronounced fruity and fermented notes. This study not only revealed the key volatile composition and differential flavor features of goji berry puree, but also provided a theoretical foundation for its quality evaluation and control, thereby offering a scientific reference for the establishment of quality standards in related industries.

To investigate the nutritional characteristics of different specialty millet varieties, this study selected the local specialty variety Qingyang Jiugu (QYJG) and the modern elite cultivar Longgu 23 (LG23) as research subjects, systematically comparing their differences in nutritional quality and metabolite profiles. The physicochemical components of both varieties were measured, and their metabolites were analyzed using widely targeted metabolomics technology. Metabolite characteristics were analyzed by principal component analysis, orthogonal partial least squares discriminant analysis, cluster analysis and metabolic pathway analysis. Results showed that the amylose content of QYJG was only 0.91%, indicating that it was a typical waxy millet variety. QYJG exhibited higher contents of unsaturated fatty acids (90.85%), polyphenols (1.00 mg GAE/g) and polysaccharides (7.66 mg/g) than LG23 (P<0.05), along with a relatively favorable fatty acid composition. In contrast, the contents of flavonoids (15.72 mg RE/g), essential amino acids (96.88 mg/kg) and total amino acids (1765.72 mg/kg) in LG23 were significantly higher than those in QYJG (P<0.01). LG23 also contained abundant levels of branched-chain amino acids and limiting amino acids, which confered a significant advantage in amino acid synthesis. A total of 1408 metabolites were identified in QYJG and LG23, primarily consisting of lipids, alkaloids, flavonoids, amino acids and their derivatives. And 587 significantly differential metabolites were screened, with 366 up-regulated in QYJG and 221 up-regulated in LG23. These differential metabolites were mainly involved in pathways related to the biosynthesis and metabolic pathways of secondary metabolites, amino acids, carbohydrates, vitamins and cofactors, as well as lipid. Lipid and carbohydrate metabolic pathways were more active in QYJG, which might contribute to its antioxidant stability and low glycemic properties. LG23 exhibited enhanced amino acid and flavonoid metabolic pathways, highlighting its superior antioxidant capacity and nitrogen metabolism efficiency. This study revealed the differences in nutritional quality and metabolic profiles between QYJG and LG23 from the perspectives of physicochemical properties and metabolomics, which provided a theoretical basis for precise breeding of millet varieties and the development and utilization of related products.

In this study, we collected Lycium barbarum L. from different areas in Ningxia and determined the contents of 17 protein amino acids. Based on the amino acid pattern recommended by FAO/WHO, we comprehensively evaluated the nutritional value of protein amino acids and regional differences using indicators, such as amino acid score (AAS), chemical score (CS), amino acid ratio coefficient (RC), and amino acid score of ratio coefficient (SRC), combined with correlation, principal component, and cluster analyses. The results revealed that Ningxia Lycium barbarum L. contained seven, nine, and five kinds of essential, medicinal, and flavor amino acids (EAA, MAA, and FAA), respectively. The total amino acid (TAA) and EAA contents ranged from 7.500 to 9.490 and 1.545 to 2.276 g/100 g, respectively. The EAA/TAA and essential-to-non-essential amino acid (EAA/NEAA) ratios were between 19.38%~26.11% and 24.03%~35.34%, respectively. Both the EAA/TAA and EAA/NTAA values were lower than the ideal protein reference values (i.e., 40 % and 60 %, respectively) given by the FAO. The total medicinal amino acid (TMAA) content ranged from 4.314 to 5.886 g/100 g, accounting for 56.59%~62.40% of the TAA, indicating high medicinal value. The total FAA content ranged from 1.557 to 2.341 g/100 g, accounting for 19.53%~26.04% of the TAA. We observed no significant differences (P>0.05) in the TAA, EAA, NEAA, MAA, and FAA contents among the different Ningxia producing areas, except for the significantly different TAA content between Tongxin and both Zhongwei and Guyuan. However, we detected significant differences (P<0.05) in the EAA/TAA and EAA/NEAA ratios as well as in the total TMAA and FAA contents. The key amino acids we identified were Gly, Asp, Glu, Met, Thr, Val, Phe, Leu, Ile, and Lys, and the combination of cysteine and methionine (Cys+Met) was the first limitation. The AAS, CS, and SRC values ranged ranged from 58.2 to 102.8, from 66.5 to 98.6, and from 71.1 to 84.9, respectively. L. barbarum L. from Zhongwei, Guyuan, Yinchuan, Haiyuan, and Zhongning regions exhibited superior comprehensive amino acid quality, with SRC evaluation results indicating a more balanced composition closer to the FAO/WHO-recommended pattern. Cluster analysis categorized the Lycium barbarum L. from Ningxia producing areas into four distinct groups: Qingtongxia and Tongxin; Yueyahu, Helan, and Hongsibao; Haiyuan and Zhongning; Guyuan, Yinchuan, and Zhongwei. The experimental results effectively reflect the regional differences in protein amino acids and the nutritional value of Ningxia Lycium barbarum L., providing a valuable reference for deep processing.